More coding requests. One predictable price.
Pick a plan by the work you expect to finish each month. Every plan serves every model, the same API formats and Z-Image Turbo images.
Lite
≈365 Kimi K3 / 30d
or ≈1,228 GLM 5.3 · ≈1,744 Qwen
5-hour pace: ≈73 Kimi · ≈246 GLM
100 Z-Image Turbo images / 30 days
Pro
≈912 Kimi K3 / 30d
or ≈3,069 GLM 5.3 · ≈4,361 Qwen
5-hour pace: ≈182 Kimi · ≈614 GLM
250 Z-Image Turbo images / 30 days
Max
≈2,280 Kimi K3 / 30d
or ≈7,673 GLM 5.3 · ≈10,902 Qwen
5-hour pace: ≈684 Kimi · ≈2,302 GLM
800 Z-Image Turbo images / 30 days
Ultra
≈4,561 Kimi K3 / 30d
or ≈15,346 GLM 5.3 · ≈21,804 Qwen
5-hour pace: ≈1,368 Kimi · ≈4,604 GLM
2,000 Z-Image Turbo images / 30 days
5 free images daily, resetting at midnight KST. Further images use $0.005 each from the same plan allowance as text, subject to the image caps above and your remaining allowance. No additional charge. Image model details
- Every served model, no tier gating
- Full published context and output limits
- Streaming, reasoning and function tools
- Per-request usage details and CSV export
How many requests does each plan include?
Estimated requests per 30 days when the whole plan is spent on one model. Each row uses a model-specific cached coding workload. Real requests vary with context length, cache hits and output size.
| Model | Lite · $10 | Pro · $20 | Max · $50 | Ultra · $100 |
|---|---|---|---|---|
| kimi-k3 | ≈365 | ≈912 | ≈2,280 | ≈4,561 |
| glm-5.3 | ≈1,228 | ≈3,069 | ≈7,673 | ≈15,346 |
| qwen-3.8-27b | ≈1,744 | ≈4,361 | ≈10,902 | ≈21,804 |
| deepseek-v4-pro | ≈1,018 | ≈2,546 | ≈6,365 | ≈12,729 |
| deepseek-v4.1-flash | ≈5,972 | ≈14,930 | ≈37,324 | ≈74,648 |
| glm-5.3-flash | ≈5,265 | ≈13,162 | ≈32,905 | ≈65,811 |
| gemma-4-31b | ≈11,165 | ≈27,913 | ≈69,782 | ≈139,563 |
| deepseek-v4-flash | ≈4,672 | ≈11,680 | ≈29,201 | ≈58,401 |
Every plan includes one shared usage allowance across the catalog. Published token rates are safety ceilings. Included usage cannot be withdrawn, and purchases are final. Refund policy
Pay by card, Apple Pay or Google Pay through Paddle, our merchant of record. No charge at signup; subscriptions renew every 30 days until cancelled, and cancelling keeps the paid term running to its end date.
What will a request actually cost?
Shape a request on the left and read its maximum charge and remaining plan allowance on the right. Completed requests use measured request usage when it is lower.
Shape a request
Set the token mix of one call. This shows the maximum charge; completed-request accounting can only reduce it.
Maximum charge per request
$0.018354
25,200 input · 800 output tokens
Requests like this on Lite
$4.03 allowance
$8.05 allowance
- InputNew content sent to the model
- $2.415
- Cache readPreviously cached input reused
- $0.2415
- Cache writeMeasured cache creation
- $2.415
- OutputGenerated response, reasoning included
- $12.075
Good to know
How do plan allowances work?
A plan includes one usage allowance shared by every model. Completed requests draw from it at the published rates. Request counts are dated estimates, not quotas.
Is image generation included?
Yes. Every plan includes Z-Image Turbo images plus five free images daily. After the daily free set, images use the published per-image rate from the shared plan allowance and count toward the plan’s image cap. There are no automatic overage charges.
Are DeepSeek V4 Pro and V4.1 Flash available?
Yes — both are included in every plan. Their rate limits may vary while capacity is tuned.
What are the allowance windows?
They are rolling windows that pace a plan. Lite and Pro allow 20% of monthly usage in 5 hours and 50% in 7 days; Max and Ultra allow 30% and 60%. Allowance returns continuously as older usage rolls off.
What is the difference between Lite, Pro, Max and Ultra?
All plans serve the same catalog and API formats. They differ in estimated monthly requests, rolling pace and price.
What happens when I hit a limit?
A plan never charges silently: when the allowance cannot cover a request you get a clear plan_usage_limit error. Upgrade the plan or wait for rolling usage to return.
How do I pay, and can I cancel?
Card payment is handled by Paddle, our merchant of record. Subscriptions renew every 30 days until you cancel; cancelling stops future renewals and the current term keeps its allowances.